We sell an SMS for market research tool. So here’s an unusual article for us to write: a list of studies you shouldn’t run by text.
We’re writing it because the fastest way to burn a new channel is to use it for the wrong job. A concept test squeezed into 160 characters produces bad data, and bad data produces skeptical clients. Then nobody wants to try texting again, even for the studies where it’s the best tool available.
So, the honest version.
1. Concept tests with visual stimuli
If respondents need to see a package design, an ad or a product mockup, SMS is the wrong home. MMS images are inconsistent across devices and carriers, can’t be controlled for size or exposure time, and can’t be rotated or randomized cleanly.
Use instead: an online survey with controlled stimulus display. SMS can still deliver the invitation link, or, RCS may be an even better option!
2. Conjoint and MaxDiff
Trade-off exercises depend on presenting multiple attributes side by side, many times over, with a design that’s carefully balanced. None of that survives a text thread.
Use instead: a dedicated online conjoint or MaxDiff tool.
3. Brand equity trackers
Trackers rely on long, consistent batteries asked the same way, wave after wave. Changing the channel changes the instrument, and changing the instrument breaks your trend line.
Use instead: keep trackers where they are. Use SMS for event-triggered pulse checks that sit alongside the tracker.
4. Long attitudinal and segmentation batteries
Twenty grid questions on a five-point agree scale is tiring on a laptop. Over text, it’s an exercise in watching completion rates collapse.
Use instead: online or email surveys for depth; SMS for the screener or the reminder.
5. Studies with complex skip logic
Branching works in SMS, but every branch is another message, and every message is another chance to drop off. If your instrument needs nested conditions or piping across many questions, the respondent experience degrades fast.
Use instead: a web survey, linked from a text if mobile reach matters.
6. Sensitive topics
Health conditions, finances, clinical experiences: texts appear on lock screens, shared devices and family plans. Respondents know that, and some will answer less honestly or not at all. In regulated settings, the compliance bar rises further, with consent and data-handling requirements beyond standard texting rules.
Use instead: a secure web survey or interview, with clear privacy framing. If you do use SMS here, use it only to deliver a secure link.
7. B2B professional panels (usually)
Many professionals see a text from an unknown number as an intrusion. Email and LinkedIn tend to fit the tone of professional research better.
Use instead: email-first outreach. Test SMS only with professionals who’ve opted in and said they prefer it.
So where does SMS for market research win?
Almost everywhere the list above doesn’t reach: short, time-sensitive questions where the moment matters more than the depth.
| SMS is a strong fit | SMS is a poor fit |
|---|---|
| Post-experience intercepts (QSR, retail, healthcare visit) | Concept tests with visuals |
| Quick 1–3 question pulse checks | Conjoint, MaxDiff, complex scales |
| Panel recruitment, reminders and re-engagement | Brand trackers and long batteries |
| Simple screeners for mobile-first audiences | Sensitive topics |
| In-the-moment diary prompts | Complex skip logic |
The pattern: SMS is for the reaction, not the reflection. When brevity is a feature of the study rather than a workaround, texting is likely the best channel you have.
A quick fit test
Before you text a study, ask:
- Can it be answered in three questions or fewer?
- Does the answer depend on when you ask?
- Can it be understood without seeing an image?
- Would a respondent be comfortable seeing it on their lock screen?
- Do you have documented SMS consent for everyone you plan to text?
Five yeses: text it. Any no: rethink the channel, or use SMS only to deliver the link.